Synergy between Interconnected Porous Carbon-Sulfur Cathode and Metallic MgB2 Interlayer as a Lithium Polysulfide Immobilizer for High-Performance Lithium-Sulfur Batteries.

Lithium-sulfur (Li-S) batteries are the potential candidates for developing high-energy-density electric vehicles. However, poor electrical conductivity of sulfur/discharged products, low active material utilization, shuttle mechanism, and poor cycle life remain the major challenges for the development of Li-S batteries. Herein, we report the nitrogen-doped highly porous carbon (NC) with interconnected pores as the sulfur host (NC-S), which is synthesized by a facile one-step process without using any template and activation agents. The highly interconnected porous structure of NC can accommodate a high amount of sulfur loading and provide space for sulfur volume expansion during redox reactions. Besides, to mitigate the lithium polysulfide dissolution and shuttle mechanism, metallic and polar magnesium diboride (MgB2) is used as an interlayer. Consequently, the NC-S/MgB2 cathode delivers higher specific capacity, rate capability, and excellent cyclic stability than the NC-S cathode and bulk sulfur cathode with MgB2 interlayer. The lithium polysulfide (LPS) adsorption test shows that MgB2 has strong chemisorption toward lithium polysulfides, which can inhibit the dissolution of LPS into the electrolyte and minimizes the shuttle effect. The dynamic electrochemical impedance spectroscopy analysis investigates the electrochemical reaction kinetics of the NC-S/MgB2 cathode during the charging and discharging processes. Overall, this work demonstrates that the synergy between the nitrogen-doped porous carbon-sulfur host and polar metallic MgB2 improves the performance of the Li-S battery, which is beneficial for the development of high-energy-density batteries for the future.

Introduction

Lithium-sulfur (Li-S) batteries are considered as the next-generation electrochemical energy storage devices owing to their high energy density (2600 Wh kg–1) and theoretical capacity (1672 mA h g–1). Besides, eco-friendliness, low cost, and wide availability of sulfur cathode make it a promising candidate.[1,2] However, the poor ionic/electronic conductivity of sulfur/discharged products, pulverization of sulfur during the redox reactions, dissolution of intermediately formed lithium polysulfides (LPS) into the electrolyte, and shuttle mechanism are the major barriers for the development of Li-S batteries in a larger scale.[3] These issues give rise to low coulombic efficiency, poor cyclic stability, and active material losses of Li-S batteries. For commercialization of Li-S batteries, the above-mentioned challenges should be addressed.

The electrical conductivity and wettability of the sulfur and discharge product lithium sulfide (Li2S) can be enriched by incorporating sulfur into carbonaceous host materials. This can improve the specific capacity, rate capability, and cyclic stability of the Li-S battery.[4,5] These parameters also depend on the choice of carbonaceous host. In this context, porous carbon nanostructures are considered as more suitable materials owing to their high specific surface area, controlled pore size distribution, and chemical and thermal stabilities.[6] Porous carbon with the combination of mesoporous and microporous structures can enhance the active materials storage and utilization and therefore, mitigate the lithium polysulfide shuttling.[7] For the synthesis of porous carbon structures, many approaches have been made using various templates. These processes are time-consuming, have complex synthesis routes, and costly.[8] Besides, addition of a nitrogen dopant to the porous carbon structure can enhance the interaction between the lithium polysulfides (LPS) and carbon structure, thereby improving the performance of the Li-S battery.[5] In the present work, a nitrogen-doped porous carbon (NC) has been synthesized by the facile thermal decomposition method. For this, sodium bicarbonate is used as pyrogen, and melamine and glucose are used as nitrogen and carbon sources, respectively. The highly porous nitrogen-doped carbon is used as a sulfur host (NC-S).

Extensive research has been carried out for developing promising polar, nonpolar, organic, and inorganic materials for the confinement of lithium polysulfides (LPS) and minimize the active material losses. The carbon-based materials can provide conductivity to the sulfur cathode and also reduce the LPS shuttling by physical confinement. This weak interaction between nonpolar carbonaceous materials and polar LPS limits the performance of the Li-S battery.[9,10] Therefore, the strong interaction between lithium polysulfides (LPS) and polar materials can trap the lithium polysulfides efficiently through chemical interactions. Transition-metal oxide, carbide, nitrides, and sulfides are mainly identified as polar materials, which can easily adsorb the LPS on their surface by the Lewis acid–base reaction, unlike nonpolar carbonaceous materials.[11,12] The shuttle mechanism is only partially limited by chemical interactions using transition-metal oxides, nitrides, carbides and sulfides due to the less number of available sites for these materials, which therefore hamper the cycling capability of the Li-S battery. To improve the performance of the Li-S battery, new materials/chemistry should be explored. The study on interactions of metal borides with lithium polysulfide has not been reported much. Li et al. have reported conductive and polar titanium boride (TiB2) as a sulfur host in Li-S batteries and effectively confined the shuttle effect.[13] Guan et al. have reported an amorphous Co2B@graphene composite for immobilization of LPS and explained that the synergistic effect of Co and B interactions with LPS and high conductivity of graphene could efficiently restrict the LPS dissolution.[14]

In this work, we report the layer-structured and room-temperature metallic nature MgB2 as an interlayer to mitigate LPS shuttling. MgB2 consists of alternative layers of magnesium and boron atoms, where the boron layer forms honeycomb stacks with the magnesium layer as a space filler.[15] MgB2 has similar band structures to graphite with deeper π bands and coexistence of 2D covalent bands from in-plane sigma bonds and 3D metallic conducting bands from the interlayer-made MgB2 as the most attractive material in the field of superconductivity.[16,17] The application of metallic, layered, and structured MgB2 as the lithium polysulfide immobilizer is not explored much. Recently, Linda F. Nazar’s group has reported ultra-lightweight MgB2 as the polysulfide mediator for LPS redox reaction.[18] They have stated that both magnesium and boron centers bind S22– anions in the absence of Li+ cations, which can enhance the diffusion of Li+ ions. The interaction reported here is different from the previously reported Lewis acid–base interactions on chalcogenides, carbides, and nitrides with LPS.[19]

In the present work, a simple one-step synthesis has been approached from the synthesis of highly porous nitrogen-doped carbon (NC) without using KOH or templates. The nitrogen-doped porous carbon is used as the sulfur host (NC-S) to fabricate the cathode. SEM, TEM, and surface area measurements confirm the highly porous nature of NC. Magnesium diboride coated on the carbon cloth with high LPS adsorption is used as the interlayer. Li-S batteries are tested with NC-S cathode with and without the MgB2 interlayer. To understand the significant performance of the NC-S cathode, the cells were fabricated using 65% sulfur, 25% carbon, and 10% binder as bulk sulfur cathode with the MgB2 interlayer and compared to the NC-S cathode with MgB2 interlayer.

Experimental Section

Synthesis of NC and NC-S

Synthesis of nitrogen-doped porous carbon (NC) involves a facile thermal decomposition technique. Briefly, melamine, glucose, and sodium bicarbonate were taken as nitrogen, carbon, and pore-forming precursors, respectively. Melamine, glucose, and sodium hydrogen carbonate were taken in the molar ratio of 1:1:1 and well ground with the help of a mortar and pestle. The ground mixture was loaded into a horizontal tubular furnace and heated at a heating rate of 5 °C min–1 to 800 °C and held at this temperature for 2 h. After the furnace temperature reaches room temperature, the sample was washed with ethanol and deionized water for the removal of sodium content and any unreacted compounds. The final NC sample was obtained by filtering and drying the washed sample at 60 °C in a vacuum oven for 10 h. The pore formation mechanism is dependent on the hydrogen-bond interactions with the material precursors. Melamine consists of molecules like N–H···N bonds arranged in planar sheets, which are connected with HCO3– anions originating from bicarbonates forming bonds like N–H···O.[20] The self-assembly of porous sheets is promoted by sodium bicarbonate.[21] The sodium bicarbonate and glucose decompose at a high temperature, liberating CO2 and H2O that lead to the formation of a porous network in the sample.[22]

Sulfur-incorporated nitrogen-doped porous carbon (NC-S) is synthesized by the melt-diffusion method. First, sublimed sulfur and NC are taken in a 4:1 ratio and a homogeneous mixture of sulfur and NC was obtained by grinding well for 1 h. The sample was sealed in an ampoule and heated to 155 °C very slowly and maintained at that temperature for 12 h and further annealed at 300 °C to vaporize the surface-deposited sulfur. Sulfur infiltration into the NC network was carried out at 155 °C, where sulfur is in a molten state with the lowest viscosity and can quickly diffuse through the pores of NC by capillary forces, which are a very slow process, and polymerization of S8 takes place.[23] The schematic illustration of the synthesis process is given in Scheme .

Scheme 1

Schematic Representation of the Synthesis Procedure for NC and NC-S

Fabrication of MgB2 Interlayer

MgB2 (Alfa Aesar, 99%, ∼100 mesh powder) is well ground with acetylene black and PVDF in an 80:10:10 ratio. The slurry was prepared by adding the NMP solvent. The uniform mixture of the slurry is coated on a carbon cloth by the doctor blade method (GDL, Nickunj Eximp Entp Pvt Ltd, India). The MgB2-coated carbon cloth is dried at 80 °C for 12 h in a vacuum oven. The dried sample is cut into a 12 mm disk and used as the MgB2 interlayer for Li-S battery.

Material Characterizations

X-ray diffraction (XRD) patterns of the samples were recorded using a Rigaku Smartlab X-ray diffractometer with Cu Kα radiation (λ = 0.154 nm) at 40 kV and 100 mA embedded with a nickel filter. The XRD patterns were recorded with a step size of 0.02°, in the range of 10–90°. The surface area analysis of the synthesized samples was studied by recording nitrogen adsorption/desorption isotherms at a liquid nitrogen temperature using a Micromeritics ASAP 2020. The specific surface area and porosity of the samples were calculated using the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) theories, respectively. The morphology of the synthesized samples was characterized by an Inspect F50 scanning electron microscopy (SEM) and a Technai G20 transmission electron microscopy (TEM) instruments. Energy-dispersive X-ray spectra analysis (EDS) measurements were studied using an Inspect F50 instrument. Thermogravimetric analysis (TGA) was carried out using an SDT Q600 from TA Instruments, at a heating rate of 20 °C min–1 from room temperature to 1000 °C. The chemical structure/bonding of the samples was studied by X-ray photoelectron spectroscopy (XPS) from a Specs X-ray photoelectron spectrometer using Al Kα enabled with PHOIBOS 100MCD analyzer as an X-ray source.

Electrochemical Measurements

The cathode of Li-S battery was prepared by mixing 75% active material (NC-S), 10% acetylene black, and 15% poly(vinylidene fluoride) binder (PVDF) with N-methyl-2-pyrrolidone (NMP) solvent in the form of a slurry. The well-prepared slurry was cast over the aluminum foil using the doctor blade technique and dried at 80 °C overnight. Similarly, bulk sulfur (S) cathode was prepared by mixing 65% sulfur, 25% carbon, and 10% binder. The dried slurry-coated aluminum foil was cut in the shape of circular disks with a 12 mm diameter and served as the working electrode. The sulfur loading was controlled as 2.5–2.8 mg cm–2. Li-S battery was fabricated in 2032 coin cells using NC-S or S electrodes as a cathode, glass fiber (GF/C) as a separator, and lithium metal as an anode. Bis(trifluoromethane)sulfonimide lithium salt (1 M, LITFSI) and LiNO3 additive (0.2 M) in 1,3-dioxolane (DOL) and 1,2-dimethoxyethane (DME) (1:1 v/v) was used as an electrolyte. The coin cell fabrication was carried out inside a glovebox (Mbraun) by maintaining O2 and H2O levels <0.1 ppm. The coin cells were assembled with and without MgB2 interlayers.

Cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), dynamic electrochemical impedance spectroscopy (DEIS) measurements were recorded using the Biologic VSP 300 electrochemical test station, and galvanostatic charge–discharge profiles were recorded using Biologic BCS-810 battery cycler. The CV and galvanostatic charge–discharge profiles were recorded in the potential range of 2.8–1.7 V. The EIS and DEIS measurements were recorded in the frequency range of 100 mHz to 1 MHz.

Lithium Polysulfide Absorption Test

Li2S6 (0.5 M) was prepared by adding lithium sulfide (Li2S) and elemental sulfur (1:5 molar ratio) in DOL and DME solvents (1:1 v/v). The suspension was stirred continuously for 72 h inside a glovebox for obtaining Li2S6. For the LPS absorption test, Li2S6 was further diluted by adding 8 μL of Li2S6 in 4 mL of DOL and DME solvents (1:1 v/v). The samples (10 mg) were added to the above solution, mixed for 30 min, and allowed to stand for 10 h. The supernatant of the samples was analyzed by UV–vis spectra using a Cary 100 UV–vis spectrophotometer from Agilent Technologies.

Results and Discussion

The X-ray diffraction (XRD) patterns for NC, NC-S, and MgB2 interlayer are shown in Figure a. The XRD pattern of NC consists of a high-intensity peak and a low-intensity peak at 26.4° and 44.1° respectively, corresponding to the (002) and (101) planes of graphitic carbon, which confirms the graphitization of porous carbon network. After sulfur incorporation into the NC structure, the XRD of NC-S depicts mainly sublimed sulfur peaks (Figure a) with a broad hump at 20–30°, suggesting the infiltration of sulfur inside the porous network of NC.[24,25] The XRD pattern of MgB2 interlayer (Figure S1) illustrates peaks at 25.2, 35.5, 42.4, 51.9, 59.9, 63.2, 66.1, 70.4, 76.1, and 83.2° corresponding to the (001), (100), (101), (002), (110), (102), (111), (200), (201), and (112) planes of MgB2 with hexagonal crystal structure (ICDD: 01-074-0982).

Figure 1

(a) X-ray diffraction pattern for NC, NC-S, and sublimed sulfur. (b) Nitrogen adsorption–desorption isotherms and (c) pore size distribution for NC.

Figure b depicts the nitrogen adsorption–desorption isotherms obtained at 77 K for NC. The sample illustrates type IV isotherm with a hysteresis loop above 0.5 relative pressure suggests the mesoporous structure of the NC. Besides, the high nitrogen uptake in the low-relative-pressure region (<0.1) indicates the presence of micropores in the sample. The specific surface area of the NC sample estimated based on the Brunauer–Emmett–Teller (BET) method is 1272.4 m2 g–1. The Barrett–Joyner–Halenda (BJH) method is used to calculate the pore volume and pore size of NC. Figure c represents the pore size distribution of NC. The sample consists of both mesopores (4 nm) and micropores (<2 nm), as indicated in Figure c inset. The NC shows a total pore volume of 1.04 cm3 g–1, including mesopore and micropore volumes of 0.77 and 0.277 cm3 g–1, respectively. The high surface area and porosity of the NC sample can enhance the electrochemical reactivity of the sulfur by providing more surface area and improved diffusion of Li+ ions and electrolytes into the porous network for the reactions to occur.

Thermogravimetric analysis (TGA) was carried out in the air atmosphere to estimate the decomposition temperature of NC and in a nitrogen atmosphere to find the sulfur weight % in the NC-S sample. Figure S2a illustrates the profile of NC from room temperature to 1000 °C in the presence of air. The weight loss below 150 °C indicates the removal of surface-adsorbed moisture. From 200 to 450 °C the NC sample wt % remains constant, and above 500 °C, the major decomposition occurs due to oxidation of carbon in NC. The TGA profile of NC-S carried out in the nitrogen atmosphere is shown in Figure S2b. The decomposition above 200 °C illustrates the oxidation of sulfur, and the final weight of sulfur in NC-S is 78 wt %. This suggests that the high surface and porous network of NC could accommodate a large amount of sulfur in it. The dominant sulfur peaks in the NC-S XRD pattern is due to the large amount of sulfur in the present sample.

The morphology of the synthesized samples is studied by scanning electron microscopy (SEM) and transmission electron microscopy. Figure a shows the SEM images for NC, where randomly distribute interconnected open pores are clearly visible. After sulfur incorporation into nitrogen-doped porous carbon, SEM of NC-S illustrates a porous network with some pores filled with sulfur nanoparticle (Figure b). The open pores of NC still remain the same. The presence of open pores, mainly mesoporous and microporous networks, can efficiently store the active material inside the cathode, therefore improving the active material utilization. The microporous configuration can accelerate the diffusion of Li+ ions, electrolyte, and electron pathways.[26]Figure c displays the TEM image of NC. The interconnected channels of the pores with uniform distribution are illustrated. The TEM image of NC-S depicts the dark sulfur particles over the porous network of the nitrogen-doped carbon, as shown in Figure d. Figure e depicts the TEM of NC-S, where the sulfur particles inside the porous carbon network are highlighted with yellow arrows. The presence of sulfur inside the nitrogen-doped porous is confirmed by energy-dispersive X-ray spectra analysis (EDS) measurements. Figure f represents the EDS pattern of NC-S, and the quantification of elements is given in the Figure f inset pie chart. The nitrogen content in NC-S is found to be 4.5%, which can also help in improving the active material utilization in Li-S battery.

Figure 2

(a, b) SEM images, (c–e) TEM images of NC and NC-S, and (f) EDS spectra inset elemental composition for NC-S.

X-ray photoelectron spectroscopy (XPS) analysis was carried out to elucidate the chemical bonding of the samples. Figure a represents the high-resolution XPS of N 1s spectra, which deconvoluted into four peaks at 398.2, 399.7, 400.7, and 402.4 eV corresponding to pyridinic-N, pyrrolic-N, graphitic-N, and oxidized-N (NO). The pyridinic, pyrrolic, and graphitic-N contents are 47.3, 28.6, and 17.1%, respectively. The high amount of pyridinic-N in the NC-S sample can bind the LPS by Lewis acid–base interactions and suppress the active material losses.[27] The high-resolution S 2p peak of NC-S is illustrated in Figure b. The S 2p spectra consist of three peaks at 163.3, 164.4, and 168.1 eV corresponding to S 2p3/2, S 2p1/2, and sulfate in air.[28,29] The binding energies of S 2p3/2 and S 2p1/2 shift to lower values than the elemental sulfur due to the interaction of carbon host and sulfur.[30] The high-resolution XPS Mg 2p and B 1s spectra are shown in Figure c,d, respectively. The Mg 2p spectra illustrate a peak centered at 50.3 eV corresponding to the metallic Mg.[31] The B 1s spectra deconvoluted into two peaks at 187.7 and 192.2 eV correspond to the zero oxidation state of boron and B2O3.[31,32]

Figure 3

High-resolution X-ray photoelectron spectra for (a) N 1s, (b) S 2p, (c) B 1s, and (d) Mg 2p.

The electrochemical performance of the synthesized materials is tested in four different configurations: (i) 65% sulfur, 25% carbon, and 10% binder cathode (S cathode) without any interlayer; (ii) 65% sulfur, 25% carbon, and 10% binder cathode with a MgB2 interlayer (S/MgB2); (iii) NC-S cathode without an interlayer; and (iv) NC-S cathode with a MgB2 interlayer (NC-S/MgB2). The electrochemical activity and the LPS adsorption capability of the S/MgB2 and NC-S/MgB2 cathodes are studied by cyclic voltammetry (CV) measurements at a scan rate of 0.1 mV s–1 in the potential range of 2.8–1.7 V. Figure a depicts the CV profiles for S/MgB2 and NC-S/MgB2 cathodes at a scan rate of 0.1 mV s–1. In the cathodic scan, the first reduction peak (I) at 2.37 V corresponds to the reduction of S8 to higher-order LPS (Li2S, 4 ≤ x ≤ 8), and the second the reduction peak (II) at 2.01 V indicates the reduction of higher-order LPS to lower-order LPS (Li2S2/Li2S). During the anodic scan, the peak at 2.4 V signifies the oxidation of lower-order LPS (Li2S2/Li2S) to higher-order LPS and sulfur. These peaks are well consistent with the galvanostatic charge–discharge profiles (Figure a&b). The NC-S/MgB2 cathode illustrates reduction peaks at 2.37 V and 2.01 V and an oxidation peak at 2.4 V. The S/MgB2 cathode displays sulfur reduction and oxidation peaks at 2.3 V, 1.91 V, and 2.45 V, respectively. The reduction peaks of the NC-S/MgB2 cathode are at more positive potentials, and the oxidation peak is more negative than the S/MgB2 cathode, suggesting the low polarization and enhanced sulfur redox kinetic of the NC-S/MgB2 cathode.[11] The anodic peak (III) and cathodic peak (I) potential differences are found to be 30 and 150 mV for NC-S/MgB2 and S/MgB2 cathodes, respectively. The lower overpotential toward the redox reaction of LPS is an effective way to reduce the cell polarization and enhance the rate capacity and durability of the Li-S battery.[33] To further understand the overpotential caused at higher scan rates, cyclic voltammetry studies were carried out at scan rates from 0.1 to 0.5 mV s–1 in the potential range of 2.8–1.7 V. The CV profiles at different scan rates for the NC-S/MgB2 and S/MgB2 cathodes are shown in Figure b,c respectively. An increase in the scan rate increases the peak current and negative shits in the potential values. The variations of the anodic peak (III) potential and the cathodic peak (I and II) potentials with respect to scan rate are given in Figure d–f, respectively. A large positive shift and negative shifts are observed in the oxidation peak (III) and reduction peaks (I and II), respectively, for the S/MgB2 cathode, which indicate the slower redox reaction kinetics toward LPS on the surface of sulfur. The addition of a nitrogen-doped porous carbon network to sulfur has significantly lowered the overpotential occurred during the lithiation/delithiation processes, suggesting the inherent LPS adsorption property.[34] This also facilitates the efficient conversion and utilization of active sulfur materials within the cathode.

Figure 4

(a) Cyclic voltammetry curves at 0.1 mV s–1 for S/MgB2 and NC-S/MgB2 cathodes; cyclic voltammetry curves for different scan rates (b) S/MgB2 and (c) NC-S/MgB2 cathodes; variation peak potential with respect to scan rate for S/MgB2 and NC-S/MgB2 cathodes (d) anodic peak (III), (e) first cathodic peak (I), and (f) second cathodic peak (II).

Figure 5

Galvanostatic charge–discharge profiles at various C rates for (a) S cathode; (b) S/MgB2, NC-S, and NC-S/MgB2 cathodes at 0.1C rate; (c) rate capability and (d) EIS profiles for S/MgB2, NC-S, and NC-S/MgB2 cathodes.

Galvanostatic charge–discharge studies were carried out at different C rates (1C-1672 mA g–1) in the potential range of 2.8–1.7 V. The galvanostatic discharge curves indicate two plateaus at 2.38 and 2.1 V corresponding to the reduction of S8 to higher-order and lower-order lithium polysulfides (LPS). In the charge curve, the plateaus indicate the oxidation Li2S to lower-order LPS, higher-order LPS, and back to S8. Figure a depicts the galvanostatic charge–discharge profiles for S cathode without any interlayer. In the first discharge cycle at 0.05C rate, the sulfur cathode delivered a specific capacity of 452 mA h g–1, and in the second cycle, the discharge specific capacity increased very little to 459 mA h g–1, which might be due to the rearrangement of sulfur atoms. However, the charge profile of the first cycles is overlapped by that of the second cycle; therefore, it is not identified clearly in Figure a. When the C rate increases to 0.1C, the specific capacity of the S cathode decreases to 342 mA h g–1, and the second plateau at 2.08 V falls to 1.95 V with a slopping line rather than plateau. This indicates the pulverization of the pure sulfur cathode and the shuttling of LPS. The galvanostatic charge–discharge profiles for the NC-S cathode are shown in Figure S3a. The NC-S cathode exhibits a specific capacity of 554, 475, 385, 298, and 156 mA h g–1 at 0.1, 0.2, 0.5, 1, and 2C rates, respectively. The NC-S cathode delivers higher specific capacity and rate capability than the S cathode, signifying the role of the porous carbon structure, which acts as a buffer layer to accommodate the volume changes occurred during the redox reactions of the sulfur.[35] Even though the specific capacity and current uptake are higher than those of pure S cathode, the unstable discharge plateaus at higher C rates indicate that the additional interlayer can improve the specific capacity of the cathode.

Figure S3b displays the charge–discharge profiles for the S/MgB2 cathode at different C rates. The specific capacities obtained by the S/MgB2 cathode at 0.1, 0.2, 0.5, 1, and 2C rates are 1015, 830, 627, 537, and 371 mA h g–1, respectively. After the addition of the MgB2 interlayer, the specific capacity values drastically improved, suggesting the strong LPS adsorption of the MgB2 surface. The Li-S battery performance of the NC-S cathode is also evaluated with the MgB2 interlayer. The galvanostatic charge–discharge profile obtained for the NC-S/MgB2 cathode from 0.1C to 2C is shown in Figure S3c. The NC-S/MgB2 cathode delivers specific capacities of 1218, 1077, 946, 809, and 624 mA h g–1 at 0.1, 0.2, 0.5, 1, and 2C rates, respectively. These specific capacity values are much higher than the NC-S cathode without interlayer and the S/MgB2 cathode, indicating that the highly porous nitrogen-doped carbon architecture with a strong LPS MgB2 adsorption mediator interlayer has significantly contributed to the confinement of the LPS shuttle effect. Therefore, the synergy between the NC-S cathode and the MgB2 interlayer delivers higher specific capacity values at all current densities. The polarization values ΔE1, ΔE2, ΔE3, and ΔE4 are measured between the linear regions of charge–discharge curves for pure S, NC-S, S/MgB2, and NC-S/MgB2 cathodes, respectively, as shown in Figure a,b. The measured polarization values for pure S, NC-S, S/MgB2, and NC-S/MgB2 cathodes are ΔE4 < ΔE3 < ΔE2 < ΔE1. Therefore, the NC-S/MgB2 cathode shows lower polarization than the other cathodes owing to the synergy of inherent polysulfide adsorption property of the MgB2 interlayer and nitrogen dopant sites in the porous carbon structure.

The capability of the Li-S battery with the NC-S, NC-S/MgB2, and S/MgB2 cathodes delivering the specific capacity at higher C rates is evaluated by rate performance studies. Figure c shows the rate performances of the NC-S, NC-S/MgB2, and S/MgB2 cathodes from 0.1 to 2C rates. The rate performance of the NC-S/MgB2 cathode is higher than the S/MgB2 cathode due to the nitrogen-doped interconnected porous structure of carbon, which can minimize the shuttling of LPS.[26] All the cathodes significantly deliver specific capacity up to 2C rate. However, the NC-S cathode without a MgB2 interlayer shows lower specific capacity, and after the addition of a MgB2 interlayer to the NC-S cathode, the specific capacity values significantly enhanced, contributing to the conductive carbon host and the polar interlayer. All the cells are cycled up to 2C rate and switched back to the 0.1C rate. The capacity retentions after switching to 0.1C for the NC-S, S/MgB2, and NC-S/MgB2 cathodes are 95.9%, 91%, and 96.6%, respectively. The NC carbon host could retain maximum specific capacity after rate performance can be attributed to the low overpotential measured from the CV profiles of the NC-S/MgB2 cathode.

The reaction kinetics of the NC-S, S/MgB2, and NC-S/MgB2 cathodes were studied by electrochemical impedance spectroscopy (EIS) analysis. The Nyquist plots for NC-S, S/MgB2, and NC-S/MgB2 cathodes after CV measurements are shown in Figure d. All the Nyquist plots comprise the depressed semicircles in the high- and mid-frequency regions and an inclined line in the low-frequency region. Nyquist plots have been fitted with an equivalent circuit model in the inset of Figure d, where Ro, Rsi, and Rct correspond to Ohmic resistance of the electrolyte, surface film resistance of the electrode, and the charge-transfer resistance between the electrode and electrolyte interface. CPE1 and CPE2 represent the constant phase elements, and CPE3 represents an infinite-length Warburg element that deals with the diffusion of ions within the cathode.[36] The Ro, Rsi, and Rct values for the NC-S, S/MgB2, and NC-S/MgB2 cathodes are given in Table . The NC-S cathode depicts a higher Rct (33.9 Ω) value than the S/MgB2 and NC-S/MgB2 cathodes, which signifies that the MgB2 interlayer can act as an upper current collector by enhancing the conductivity of the cathode.[37] Besides, the LPS adsorption property of MgB2 also enhances the redox kinetics of sulfur and promotes the high utilization of active materials. The NC-S/MgB2 cathode shows lower Ro, Rsi, and Rct values compared to the S/MgB2 cathode, suggesting that the synergistic combination of nitrogen-doped highly porous carbon network and MgB2 interlayer can significantly improve the active material utilization of Li-S battery and inhibit the shuttling mechanism.

Table 1

Ro, Rsi, and Rct Values for NC-S, S/MgB2, and NC-S/MgB2 Cathodes

cathode	Ro (Ω)	Rsi (Ω)	Rct (Ω)
NC-S	5.1	30.7	33.9
S/MgB2	2.8	7.1	5.1
NC-S/MgB2	2.2	5.5	3.9

The long-term cyclic stability of the NC-S, S/MgB2, and NC-S/MgB2 cathodes are evaluated at 1C rate in the potential window 2.8–1.7 V for 500 cycles (Figure ). The NC-S without any interlayer could show cyclic performance up to 500 cycles, while the capacity value is less. At the end of 500 cycles, the NC-S cathode could retain 66.4% of its initial capacity. S/MgB2 cyclic stability curves indicate the decrease in the specific capacity values with increasing number of cycles. The capacity retention of the S/MgB2 cathode after 500 cycles is 71%. Even though MgB2 has strong adsorption toward LPS, the poor stability of the sulfur cathode degrades the performance of the S/MgB2 cathode. The cyclic performance of the NC-S/MgB2 cathode is illustrated in Figure . The NC-S/MgB2 cathode could retain 85% of its initial capacity after 500 cycles, with nearly 99% coulombic efficiency. This suggests the synergistic effect of highly porous nitrogen-doped carbon host and layered metallic MgB2. As reported in the literature, Mg and B atoms can adsorb polysulfide anions strongly and facilitate Li+-ion diffusion without binding to Mg and B atoms.[18] This has also contributed to the enhanced cyclic stability and rate capability of the MgB2 interlayer. Nitrogen doping of the carbon lattice, mainly pyridinic-N and graphitic-N, can strongly bind the LPS via chemisorption and minimize the active material losses.[27] Therefore, combination of the materials of these two properties, such as the NC-S/MgB2 cathode for Li-S battery, results in an increase of specific capacity, rate performance, and cyclic stability of the battery.

Figure 6

Cyclic stability at 1C rate for NC-S, S/MgB2, and NC-S/MgB2 cathodes.

To elucidate the electrochemical process taking place on the NC-S/MgB2 cathode during the charging and discharging processes distinctly, the dynamic electrochemical impedance spectroscopy (DEIS) technique has been studied. The DEIS technique is beneficial over EIS, where the AC signal is superimposed by the DC signal during the charging/discharging process, and the measurements are recorded in the quasi-stationary state.[38,39] This technique is helpful in studying the electrochemical reaction kinetics of the electrode materials in both charge and discharge states separately. Figure a,b depicts the DEIS profiles of the NC-S/MgB2 cathode during the discharging and charging processes, respectively. Similar to EIS, the Nyquist plots of DEIS also comprise depressed semicircles in high- and mid-frequency regions, followed by a slopping line in the low-frequency region. It is clearly visible from the DEIS profile of the NC-S/MgB2 cathode during the discharging process that with increasing depth of discharge (DOD), the length of the diffusion tail decreases until the voltage reaches 2.36 V. This is because as the voltage reaches from 2.8 to 2.36 V, S8 reduces for the formation of higher-order LPS (Li2S8); during this stage, Li+ ions have faster diffusion kinetics and the diffusion tail decreases. Above 2.36 V, the higher-order LPS reduces for the formation of lower-order LPS, which increases the viscosity of the electrolyte, and therefore, diffusion tails increase.[23] During the charging process, a similar trend in the DEIS profile has been observed. The length of the diffusion is minimum around 2.36 V due to the formation of higher-order LPS.[40,41]

Figure 7

DEIS profiles for the NC-S/MgB2 cathode during (a) discharging and (b) charging processes. Resistance versus voltage profile during (c) discharging and (d) charging processes.

The electrochemical reaction kinetics of the NC-S/MgB2 cathode are analyzed by fitting the Nyquist plots during the discharging and charging processes (Figure a,b using the equivalent circuit given in the inset of Figure d). The variation of Ro, Rsi, and Rct with respect to DOD is illustrated in Figure c. The Ohmic resistance (Ro) values remain uniform during the discharging process, which can be attributed to the synergistic combination of the NC sulfur cathode and the metallic MgB2 interlayer that can inhibit LPS dissolution into the electrolyte. The Rsi value is constant at the beginning of the discharging process and it decreases with increasing DOD up to 2.2 V due to the formation of higher-order lithium polysulfides on the surface of the MgB2 interlayer, which is mostly in liquid state and decreases the surface film resistance. Above 2.1 V, Rsi increases for the conversion of higher-order LPS to Li2S2/Li2S films on the cathode surface. The Rct value is higher during the initial discharging process due to the presence of insoluble sulfur spices; as DOD increases, the Rct value decreases for the conversion of S8 to Li2S8/Li2S6 with faster reaction kinetics on the MgB2 interlayer. Above 2.2 V, Rct tends to increase due to an increase in the concentration of insoluble and nonconductive LPS (Li2S2/Li2S) species on the cathode. The variation of resistance values (Ro, Rsi, and Rct) during the charging process is represented in Figure d. Similar to the discharging process, the Ro value remains constant during the charging process of NC-S/MgB2 cathode. During the initial charging condition, the Rsi value is higher due to the presence of Li2S2/Li2S on the cathode surface and decreases as the conversion of lower-order LPS to higher-order LPS proceeds. At the end of charging, the Rsi value increases due to the deposition of a sulfur film on the cathode. The Rsi value represents the resistance offered due to the diffusion of electrolyte through pores of the interlayer, which suggests that the electrochemical process taking place during the charging is governed by the transport properties of the electrolyte.[42] During charging, the Rct values tend to decrease from 2.2 to 2.5 V due to the formation of higher-order LPS and reaches maximum at the end of the charging process.

Evaluation of Lithium Polysulfide Confinement

The adsorption capability of nitrogen-doped porous carbon (NC) and MgB2 toward LPS is evaluated by the LPS adsorption test. For this, NC and MgB2 are added in a blank Li2S6 solution, mixed well, and kept undisturbed for 10 h inside an argon-filled glovebox. Figure a inset displays the digital photographs of blank Li2S6 solution and Li2S6 solution with NC and MgB2 samples. Compared to the blank Li2S6 solution, the NC sample shows a very light yellowish color due to the adsorption of LPS. The MgB2 sample solution becomes completely transparent, suggesting the strong adsorption of LPS on its surface. This is further analyzed using UV–vis photospectrometer by collecting the supernatant solution of NC and MgB2 samples. A UV–vis spectrum is obtained for the blank Li2S6 solution for comparison. In Figure a, the UV–vis spectra of the blank Li2S6 solution show a peak at ∼357 nm corresponding to the absorbance of S62– anions.[43] The peak intensity reduces for the supernatant solution NC, indicating the adsorption of LPS on the surface but not completely. On the other hand, for the MgB2 supernatant solution, the peak at ∼357 nm disappears, indicating the absence of Li2S6 in the solution. Moreover, the intensity of absorbance also reduces for the supernatant solution of MgB2 compared to NC and blank Li2S6 solution. This also corroborates the strong adsorption capability of MgB2 toward LPS, which can minimize the shuttle effect during cycling. This also infers that polar metallic compounds have strong LPS adsorption capability over carbonaceous materials. However, the carbonaceous material as the sulfur host can accommodate a large amount of sulfur loading and suppress the volume changes during the redox cycle due to its high surface and porosity.

Figure 8

(a) UV–vis spectra and (inset) digital photograph of LPS adsorption of NC, MgB2, and reference blank Li2S6 solution; NC-S cathode (b) uncycled, (c) cycled with MgB2 interlayer, (d) uncycled MgB2 interlayer, (e) cycled MgB2 interlayer with NC-S, and (f) cycled MgB2 interlayer with S cathode.

To understand the interaction of LPS with electrode and interlayers, the morphology of the cycled electrode and interlayers and uncycled electrodes are studied SEM images. Figure b illustrates the SEM image of the uncycled NC-S electrode with interconnected open pores. After cycling the NC-S electrode, the pores of nitrogen-doped porous carbon covered the polysulfides, suggesting the porous structure of NC contributed to anchoring the LPS in the cathode (Figure c). The SEM images of uncycled MgB2 interlayer and cycled interlayer with the NC-S cathode and pure sulfur cathode are represented in Figure d–f, respectively. The morphology of the cycled MgB2 interlayer with the NC-S cathode is changed when compared to the uncycled one where larger and discrete particles of the MgB2 interlayer are identified. This can be due to the adsorption of LPS on its surface. The cycled MgB2 interlayer with S cathode shows the high polysulfide deposition of LPS on the surface of MgB2 with very less discrete particle nature. This can be attributed to the poor adsorption capability of the sulfur cathode.

Conclusions

In summary, nitrogen-doped porous carbon (NC) with interconnected pores has been synthesized by a facile one-step synthesis method and used as the sulfur host (NC-S) for Li-S battery. The high surface area for NC can accommodate a high amount of sulfur loading in the cathode and also provide space for the sulfur volume change and prevent pulverization. The NC-S cathode delivers higher specific capacity (554 mA h g–1 at 0.1C) and rate capability (up to 2C) than bulk sulfur (S) cathode, signifying the role the porous of the carbon structure. To prevent the polysulfide dissolution and inhibit the active material losses, room-temperature metallic and polar material MgB2 has been used as an interlayer. The sulfur cathode with MgB2 interlayer (S/MgB2) delivers a specific capacity of 1015 mA h g–1 at 0.1C rate, much higher than the sulfur cathode, owing to the inherent LPS adsorption property of MgB2. The NC-S cathode with MgB2 interlayer (NC-S/MgB2) delivers a specific capacity of 1218 mA h g–1 at 0.1C rate higher than the S/MgB2 cathode. Moreover, the NC-S/MgB2 cathode shows excellent cycling stability up to 500 cycles, with 85% of capacity retention. The high performance of NC-S/MgB2 can be attributed to the synergy between the interconnected porous network of the NC cathode and the strong LPS adsorption capability of the MgB2 interlayer. The dynamic electrochemical impedance spectroscopy (DEIS) analysis elucidates the electrochemical reaction kinetics of the NC-S/MgB2 cathode during the charging and discharging processes. In conclusion, the synergy of the lithium polysulfide confinement and conversion reactions of the NC-S/MgB2 cathode paves the way for the development of high-energy-density lithium-sulfur batteries.